1 Citation: Yang, C., Li, R., Cui, C., Liu, S., Qiu, Q., Ding, Y., Wu, Y., and Zhang, B. (2016). Catalytic hydroprocessing of microalgae-derived biofuels: a review. Green Chemistry, 18(13), 3684-3699. DOI: 10.1039/c6gc01239f Web: http://pubs.rsc.org/en/content/articlelanding/2016/gc/c6gc01239f/ Catalytic Hydroprocessing of Microalgae-Derived Biofuels: A Review Changyan Yang 1,2 , Rui Li 3 , Chang Cui 1 , Shengpeng Liu 1 , Qi Qiu 4 , Yigang Ding 1 , Yuanxin Wu 1 , Bo Zhang 1 1 School of Chemical Engineering and Pharmacy, Wuhan Institute of Technology, Hubei, China 2 Hubei Key Laboratory for Processing and Application of Catalytic Materials, Huanggang Normal University, Hubei, China 3 Joint School of Nanoscience and Nanoengineering, North Carolina A & T State University, Greensboro, NC, United States 4 College of Chemistry and Environmental Engineering, Shenzhen University, Guangdong, China Abstract The algal biofuel technology has been accelerated greatly during last decade. Microalgae can be processed into a broad spectrum of biofuel precursors, which mainly include crude algal oil recovered by extraction and bio-crude oils produced from hydrothermal liquefaction and pyrolysis processes. Due to the high protein content in algal species and the limitations of conversion technologies, these biofuel precursors require further catalytic removal of heteroatoms such as oxygen, nitrogen, and sulfur, being upgraded to biofuels like green diesel and aviation fuel. This article reviews the state-of-the-art in hydroprocessing of microalgae- based biofuels, as well as the catalyst development and the effect of process parameters on hydrotreated algal fuels. Hydroprocessing of algal fuels is a new and challenging task, and still underdeveloped. For the long term, an ideal catalyst for this process should possess following characteristics: high activities towards deoxygenation and denitrogenation, strong resistance to poisons, minimized leaching problems and coke formation, and an economically sound preparation process. Keywords: Microalgae-derived Biofuels; Hydroprocessing; Hydrodenitrogenation (HDN); Hydrodeoxygenation (HDO); Catalyst Development; Process Parameters Contents 1. Introduction ................................................................................................................................... 2 2. Overview of Microalgae-Derived Fuels........................................................................................ 5 2.1 Algal Oil Recovered by Extraction and Algae-Based Biodiesel ............................................. 5 2.2 Bio-Crude Oil via Hydrothermal Liquefaction ....................................................................... 6
28
Embed
Catalytic Hydroprocessing of Microalgae-Derived Biofuels ...thefutureenergy.org/wp-content/uploads/2016/06/Catalytic... · Catalytic hydroprocessing of microalgae-derived ... Catalytic
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Citation: Yang, C., Li, R., Cui, C., Liu, S., Qiu, Q., Ding, Y., Wu, Y., and
Zhang, B. (2016). Catalytic hydroprocessing of microalgae-derived biofuels: a
review. Green Chemistry, 18(13), 3684-3699. DOI: 10.1039/c6gc01239f
With growing concerns about declining fossil fuel supplies, environmental issues,
and increasing demand of fossil fuels, renewable biofuels have received a large amount
of research attention. While the first generation sugar or oil based biofuels (i.e. ethanol
and biodiesel) cannot meet the requirement for fuel supply and caused a conflict
between food and fuel production, the second generation advanced biofuels (such as
cellulosic ethanol and cellulosic butanol) are still under-development and gradually
entering market 1. Recently, algae were considered as a promising third-generation
biofuel feedstock due to their superior productivity, high oil content, and
environmentally friendly nature 2. Algae perform oxygenic photosynthesis like higher
plants, representing a big variety of species living in a wide range of environmental
conditions 3. Algae are not traditional foods or feeds, and they can be cultivated in large
open ponds or in closed photobioreactors located on non-arable land. Some algal
species hold higher potential as the oil-producer than oil crops. Algae can sequester
carbon (CO2) from many sources and may be processed into a broad spectrum of
products including biodiesel, green diesel, gasoline replacements, bioethanol, methane,
heat, bio-oil, fertilizer, high protein animal feed, etc. 4
3
The algal technology for biofuels production has been greatly advanced over the
past decade 5. Experts from industry, academia, and national laboratories made
invaluable contributions to its development from the biology to fuel conversion,
reducing the cost of algae-based bio-crude from $240 to $7.50 per gallon 6. However,
in order to meet the long term goal of $3/gasoline gallon equivalent, it still requires a
combination of improvements in all key technologies including productivity,
conversion, and processing 7.
As shown in Figure 1, currently, there are three approaches that are used mainly
for producing algae-based biofuels. The first technique involves extracting lipids from
algal cells, which is followed by transesterification of triglycerides and alcohol into
fatty acid alkyl esters (i.e. biodiesel) 8 or upgrading (i.e. algal lipid upgrading, also
called ALU pathway) 9. The second technique employs the hydrothermal liquefaction
(HTL) process that produces water-insoluble bio-crude oil (simply called bio-oil) by
using treatments at high pressure (5-20 MPa) and at the temperature range of 250-450°C 10. Bio-oil produced after the water separation has lower water content and thus higher
energy content than that produced directly by biomass pyrolysis. The third technique
relies on the pyrolysis technology, which thermally degrades biomass at 300-700°C in
the absence of oxygen, resulting in the production of bio-oils, solid residues, and
gaseous products. The advantages of this technique include short process time,
increased process yield, and environmental compatibility 11, 12. Both pyrolysis and
traditional lipid extraction might not be practical for algal biomass due to its high water
content. The dehydration is energy prohibitive, which limits the options for algae as
feedstock and overall process economy 13, 14. Thus, the ALU process and the HTL
process are chosen by U.S. Department of Energy (US DOE) as the two most promising
approaches 15-17.
Figure 1. Strategies for fuel production from algae
Other techniques for converting algae to biofuels include gasification (supercritical
4
water or steam) 18, 19 and biological conversion of sugar-rich algae 20. The products from
these two processes are hydrogen and ethanol, respectively. It will require extensive
efforts prior to bringing up more research interests on these two processes. Usually, the
choice of the conversion technology is dependent on the composition of available
feedstock. For example, biological conversion is preferred for marine macroalgae with
the high carbohydrate content 20, while HTL uses the whole algae 21, although their
biochemical makeup has important effects on the yields and the product distribution 22.
Table 1. Composition of microalgal biomass used for biofuel production
Nannochloropsis sp. 61, residues after lipid extraction 59, 62, and oleaginous algal species 63. Recent developments of algal pyrolysis research have been reviewed by Marcilla
et al. 64, and Brennan and Owende 65.
Pyrolysis of algae yields three streams of products (i.e. condensed liquid, gaseous
products, and biochar). In most publications, this liquid is called bio-oil. Because a
pyrolytic bio-oil normally contains 30-50% water, it will simultaneously form two
layers of products: water phase and oily phase, which were called aqueous products (or
water solubles) and bio-oil, respectively 28. The product yields for bio-oil, water
solubles, and gases are in ranges of 18-57.9%, 15-30%, and 10-60%, respectively 64.
The problems of algal pyrolytic bio-oils are similar to those of HTL oils and
lignocellulosic biomass-based pyrolytic bio-oils. A comparison of properties of HTL
and pyrolysis bio-oils is shown in Table 2. The high oxygen content in the pyrolytic
bio-oil caused low vapor pressure, low heating value, and low thermal stability. In
addition, because the high protein content in almost all algal species, the nitrogen
content in pyrolytic bio-oil is somewhere between 5-13%. Thus, in order to apply algal
bio-oils as the transportation fuel, it’ll require reduction of both nitrogen and oxygen
contents.
Table 2. Ultimate Analysis of HTL and Pyrolytic Bio-Crude Oils
The hydrotreating experiments resulted in a nearly complete conversion of microalgal
oil and a 56.2% yield of hydrocarbons with a range of C13-C20. But the activity of the
sulfided NiMo/γ-Al2O3 only lasted for 7-h before deactivation due to coke formation.
Compared with noble metal catalysts (Pt/Al2O3 and Rh/Al2O3), the NiMo bimetallic
catalyst formed less coke and required less H2 than noble metals 81.
Iron is also an attractive candidate for this kind of conversion due to low cost, rich
redox-chemistry, and high natural reserves. One study explored the possibility of
using iron nanoparticles supported on mesoporous silica nanomaterials (Fe-MSN) in
hydrotreating reactions 82. The merit of the Fe-silica catalyst is the high selectivity for
hydrodeoxygenation over cracking and decarbonylation. But the hydrotreatment over
this Fe-silica catalyst only gave a 67% conversion, and the products contained a
significant amount of alcohols (16%).
With regard to hydrogenation of biodiesel, most previous researches were
conducted using vegetable oil or other feedstock based biodiesel 85-87. Since the nature
of different biodiesels is similar, their conclusions might be applicable to algae-based
biodiesel. Recently, a report studied hydrogenation of algae-based biodiesel in
dodecane over 5 wt% Pd/C and 5 wt% Ni/HY-80 (SiO2/Al2O3=80). Hydrotreating algal
biodiesel was done at 300°C and 3 MPa H2. The performance of Ni/HY-80 was superior
to Pd/C catalyst, giving a ~95% yield of hydrocarbons that mainly comprised
octadecane, hexadecane, and heptadecane 88.
3.2 Catalytic Hydroprocessing of Algal HTL Bio-Crude Oil
Developments in the field of catalytic hydroprocessing of cellulosic biomass-
derived liquefaction bio-oil between 1980-2007 have been documented by Elliott 89.
Recent studies on hydroprocessing of algal HTL bio-oils are summarized in Table 4.
Hydroprocessing of algae-derived fuels differs from upgrading lignocellulosic
biomass-derived oils because of the importance of both deoxygenation and
denitrogenation. Thus, an algal bio-oil upgrading process needs to fulfill following
12
purposes: oxygen and nitrogen removal, molecular weight reduction, minimizing
hydrogen consumption, and avoiding saturation of the aromatic rings.
3.2.1 Hydroprocessing Bio-Oil with Molybdenum based and Noble Metal
Catalysts
The work of hydroprocessing of the HTL bio-crude oil, led by Pacific Northwest
National Laboratory (PNNL), examined allover material balances and upgraded fuel
quality. Their hydroprocessing experiments were performed in a two-stage continuous
system 90 by using sulfided CoMo/F-Al2O3 (KF-1001, Akzo Chemicals Inc. 91). A
relatively high upgraded oil yield of 80-85% was obtained, and they concluded that this
hydroprocessing process was effective for deoxygenation, denitrogenation, and
desulfurization of the bio-oil from Nannochloropsis alga 92. The oxygen, nitrogen, and
sulfur contents in algal bio-oils were reduced to 1-2%, <0.5%, and <50 ppm,
respectively. The products in the upgraded bio-oil fell primarily in the diesel range. The
similar results were also confirmed for Chlorella alga, as the oxygen, nitrogen, and
sulfur contents in algal bio-oils were reduced to 2.2%, <0.05%, and <50 ppm,
respectively 7. Further, the same process was applied to Chlorella grown
heterotrophically, which had a lipid content of 57–64% and low nitrogen content of 0.5% 93. After a hydrotreatment, the oxygen, nitrogen, and sulfur contents of this algal biofuel
were reduced to 1.7%, <0.05%, and 18 ppm, respectively.
Concerning both the yield of treated HTL oil and the effects of HDO, HDN, and
HDS, the results obtained by researchers at the PNNL are remarkably better than those
of rest studies that are reviewed in following sections. The possible reasons are twofold:
high availability of H2; and the reaction by-products (such as NH3, H2O, H2S, and cokes)
were removed immediately from the continuous process, minimalizing their windows
for reacting with hydrocarbons to form undesired products.
A similar work was done by scientists at the University of Leeds and the University
of Illinois 66, who hydroprocessed bio-crude oil from hydrothermal liquefaction of
Chlorella. Both non-catalytic and catalytic hydroprocessing reduced nitrogen and
oxygen contents in the upgraded oil, giving an oil yield between 41-94.8%. The treated
oil can be fractionated into 25% gasoline, 50% diesel, and 25% heavy fuel oil. However,
the lowest N content reached in this study was 2.4% by using NiMo/Al2O3 at 405°C,
so the catalytic function of catalysts towards hydrodenitrogenation needs to be further
improved. Authors also pointed out the differences between their work and the PNNL
study: 1) the higher O and N contents in the bio-crude (11-16% O and 6% N) compared
to PNNL bio-oil (5-8% O and 4-5% N); and 2) hydroprocessing was conducted at the
batch mode.
Some recent studies compared the catalytic effects of noble metals (like Pt and Ru)
with transition metals (Ni, Mo, etc.) using various species of algae 94, 95. Noble metals
showed higher HDO activities than transition metals. But even the amount of Ru/C was
increased to 30% of the total loading, the N content in upgraded oils was still higher
than 2.4% of total oil 95. In addition, the noble metals were often deactivated within a
short period of time. The selection of the active metal for hydroprocessing bio-crude
oil is discussed in the section 4.
13
Table 4 Studies on catalytic hydroprocessing of algal HTL bio-crude oil
Microalgae
provider or
species
Catalysts Experimental details Key results Ref
Nannochloropsis sulfided CoMo/F-
Al2O3 (4% Co and
15% Mo)
Hydroprocessing was
conducted in a bench-scale (412
mL), two-stage continuous
system. The operation
conditions for first and second
stages were (125-170°C and
13.6 MPa) and (405°C and 13.6
MPa), respectively.
They obtained an upgraded oil yield of 80-
85% The products in the upgraded bio-oil
that had a carbon number range of C6-C32
fell primarily in the diesel range (C14-
C18).
92
Chlorella sulfided CoMo/F-
Al2O3
Same as the above The oxygen, nitrogen, and sulfur contents
in algal bio-oils were reduced to 2.2%,
<0.05%, and <50 ppm, respectively.
7
Chlorella grown
heterotrophically
sulfided CoMo/F-
Al2O3
The bio-oil was upgraded at
400°C and 10.3 MPa (1500 psi)
H2 in a continuous system
Compared with the phototrophic culture,
this alga produced twice amount of bio-oil
and upgraded oil.
93
Chlorella sulfided
NiMo/Al2O3 or
CoMo/Al2O3
The hydroprocessing was
operated at 350°C or 405°C
under 6-6.6 MPa of initial H2
pressure in a 500 ml Parr
reactor.
The upgraded oil yield was between 41-
94.8%. The treated oil contained alkane
hydrocarbons ranging from C9 to C26, and
can be fractionated into 25% gasoline, 50%
diesel, and 25% heavy fuel oil.
66
Scenedesmus sp. Pt/C, Ru/C, Ni/C,
and Co/C
The experiments were carried
out at 350°C under 6.9 MPa of
initial H2 pressure for 4 h in a
450 mL Parr reactor
Ru/C and Pt/C had the best efficiency in
hydrogenation, and enhanced the
production of octadecane and hexadecane.
94
Spirulina
platensis,
Nannochloropsis
sp., and a mixture
of Chlorella
sorokiniana,
Chlorella
minutissima, and
Scenedesmus
bijuga,
Ru/C and sulfided
CoMo/Al2O3
Hydrodeoxygenation was
performed at 350°C under 5.17
MPa of H2 pressure for 4 h in a
batch reactor.
HDO reduced nitrogen heteroatoms in bio
crude oil to 2.4.-3.1%.
95
Nannochloropsis
sp.
Pt/C in the
presence of water
The hydrotreating experiments
were performed by adding
certain amount of water in HTL
bio-oils of Nannochloropsis sp.,
which were followed by
treatments in a 4 mL mini-
Pt/C resulted in an oil yield of 77% and
82% carbon recovery. However, the N and
O contents in treated oils were still in
ranges of 1.99-3.98% and 3.08-6.97%,
respectively.
96
14
reactor at 400°C and 3.4 MPa H2
for 1-4 h.
Nannochloropsis
sp.
Pd/C Same as the above The use of Pd/C produced oils with 44
MJ/kg HHV and a yield of 79%. The most
abundant alkane in the treated oil was
pentadecane (C15) coexisting with others
ranging from C8 to C32.
.
97
Nannochloropsis
sp.
Pt/C, Mo2C, and
HZSM-5
HTL oil upgrading was carried
out in a stainless-steel mini
batch reactor with 0.5 g of crude
bio-oil, the desired amount of
catalyst, and 0.4 ml water.
Factors of temperature (330-
530°C), time (2-6 h), catalyst
types, and catalyst loading (5-20
wt%) were varied.
The reaction temperature was the most
influential factor. The most abundant
alkane in the treated oils was pentadecane
(C15), and others alkanes ranging from
C10 to C31 are also present.
98
Chlorella
pyrenoidosa
Pt/γ-Al2O3 HTL oil upgrading was done in
supercritical water (400°C) for 1
h, and H2 or formic acid was
used as the source of electrons.
Under supercritical water conditions,
reactions caused an oil yield 60-70%. GC-
MS showed the treated oil contained a
series of n-alkanes starting at C11.
99
Chlorella
pyrenoidosa
Pt/C, Pd/C, Ru/C,
sulfided Pt/C,
Pt/C(CO), Pt/C(n-
C6H14), Mo2C,
MoS2, Al, sulfided
CoMo/γ-Al2O3,
Ni/SiO2–Al2O3,
HZSM-5,
activated carbon,
and Al/Ni
The hydrotreatment was done at
400°C and 6 MPa H2 in a 58 mL
reactor filled with 3 g bio-oil,
0.3 g catalyst, and 1.5 mL water.
The process showed upgraded oil yields of
53.1-77.2%. When using Ru/C with Raney
Ni as the catalysts, the upgraded oil flows
freely, and has 97 wt% of the material
boiling below 400°C and a heating value of
45 MJ/kg.
100
Chlorella
pyrenoidosa
A mixture of Ru/C
with one of above
mentioned
catalysts
The hydrotreatment was
performed at 400°C for 4 h in a
batch reactor. For each run, 3 g
bio-oil, 0.3 g catalyst (0.15
Ru/C and 0.15 g other catalyst),
1.5 mL water, and 6 MPa H2
were loaded into the reactor.
Ru/C & Mo2C produced the highest oil
yield of 77.2% and energy recovery. The
treated bio-oil contained straight-chain
alkanes ranged from C10 to C25, with
pentadecane (C15), hexadecane (C16), and
heptadecane (C17) as the three most
abundant hydrocarbons.
23
3.2.2 Hydrotreating Bio-Oil in the Presence of Water
A series of studies on hydrotreating of bio-oil in supercritical water were conducted
by Dr. P.E. Savage and his collaborators at the Henan Polytechnic University. The
15
motivation of their research is from a process engineering perspective, to take
advantages of hydroprocessing HTL bio-oil in the same environment as HTL 97. Their
initial hydrotreating experiments were performed by adding certain amount of water in
HTL bio-oils of Nannochloropsis sp., which were followed by hydrotreatments with
Pt/C 96 and Pd/C 97 catalysts. The use of Pd/C produced oil with 44 MJ/kg HHV and a
yield of 79%, while Pt/C resulted in an oil yield of 77% and 82% carbon recovery.
However, the N and O contents in treated oils were still in ranges of 1.99-3.98% and
3.08-6.97%, respectively. Further, they compared the functions of three catalysts: Pt/C,
Mo2C, and HZSM-5, and concluded that reaction temperature was the most influential
factor 98. Among the catalysts studied, applying Mo2C at 530°C for 2 h showed the best
deoxygenation performance, and using Pt/C at 530°C for 6 h resulted in the lowest N
content of 1.5% in the treated oil.
Their recent research subject changed to HTL oils of Chlorella pyrenoidosa, which
has a lower O content of 2.1-7.8% but higher N content of 7.8-8.0%. Treatment of this
bio-oil was done with Pt/γ-Al2O3, varying the source of electrons: H2 or formic acid 99.
Although this research indicated that 0.025 g/cm3 water density is the optimal condition
for hydrotreating bio-oils, both deoxygenation and denitrogenation functions of Pt/γ-
Al2O3 were not effective under supercritical water conditions.
Later on, fifteen hydrogenation catalysts including Pt/C, Pd/C, Ru/C, sulfided Pt/C,
HZSM-5, activated carbon, and Al/Ni were tested for upgrading HTL oils of C.
pyrenoidosa 100. The yields of upgraded oil fell in the range of 53.1-77.2%. Ru/C gave
the best result for deoxygenation, and Al/Ni (Raney nickel) was shown to be a suitable
catalyst for denitrogenation. Catalysts of Co-Mo/Al2O3, Mo2C, and MoS2 performed
poorly for deoxygenation in the presence of water, but remained high denitrogenation
activity that is comparable to that of the noble metal catalysts.
In their recent work, a mixture of Ru/C with one of above mentioned catalysts was
used. In respect of deoxygenation, denitrogenation, and desulfurization, Ru/C & Mo2C,
Ru/C & Pt/γ-Al2O3, or Ru/C & Pt/C showed the best results, giving the O, N, and S
contents of 0.1, 1.8, and 0.065 wt%, respectively 23. Ru/C & Mo2C produced the highest
oil yield of 77.2% and energy recovery. Although these experiments were performed in
an engineering way, the results revealed some insights in hydrotreating reactions,
indicating that catalytic synergy in bimetallic catalysts is worth further research.
However, the N content of upgraded oils could not be reduced to less than 1.5% under
supercritical water conditions according to their reports. Therefore, both the
hydrotreating process and the catalyst will require further improvements.
3.3 Catalytic Hydrotreating of Algal Pyrolytic Bio-oil
As a separate unit operation, the study of hydrotreating always followed the waves
of developments of algal research or conversion technologies. Even though there are an
increased number of algal pyrolysis studies since 2009, up-to-date, only few articles
reported hydrotreating of algal pyrolytic bio-oil. Zhong et al. studied hydrotreating of
fast pyrolysis oil from Chlorella over a Ni-Co-Pd/γ-Al2O3 catalyst 101. Hydroprocessing
16
at 300°C and 2 MPa H2 resulted in a refined oil yield of 89.6% and an 80.4% reduction
of the oxygen content. The nitrogen content was reduced from 6.48% to 2.45%.
In another study, bimetallic Ni-Cu/ZrO2 catalysts with various Cu/Ni ratios (0.14
to 1.00 w/w) were used for HDO of pyrolytic bio-oils of Chlorella sp. and
Nannochloropsis sp. at 350°C and 2 MPa H2 25. Compared with Ni/ZrO2 and sulfided
NiMo/Al2O3, the addition of copper could facilitate the reduction of nickel oxide and
limit sintering and coking, showing a higher HDO efficiency of 82%. But the
denitrogenation activities of catalysts were not even considered.
4. Catalyst Development for Hydrodenitrogenation of Algal Bio-Oil
The crude algal oil produced from the ALU pathway has low S and N contents, so
the traditional HDS catalysts might be effective enough to remove oxygen and improve
its quality. A detailed review of the catalyst design strategies for hydrodeoxygenation
can be found at Arun et al. 74. For upgrading algal HTL and pyrolytic bio-oils, the
catalysts have to be bifunctional, possessing both HDO and HDN activities. Until now,
a limited number of catalysts have been investigated, and the results showed both
promising possibilities and significant problems.
Most hydrogenation catalysts could denitrogenate the algal fuels to some extent,
but the nitrogen content left in the hydrotreated oil was often between 1% and 4%. The
residual nitrogen-containing compounds are in forms of pyrrole 101, amides (like N,N-
dimethylhexanamide, palmitamide, benzenamine), nitriles, quinolone 96, and indole 98.
Although ASTM and EN standards do not regulate the minimal nitrogen content of
current transportation fuels, US DOE’s goal for the nitrogen content in upgraded algal
fuels is less than 0.05% (500 ppm) 17. In order to meet this goal, further development
of highly selective catalysts for the C-N bond breakage is needed.
A significant number of HDN studies have been done with model chemicals.
However, because the algal fuels are complex mixtures, the most active catalyst for
HDN of model compounds might not show the highest catalytic activity for upgrading
algal bio-oils 100. Conversely, the use of model compounds is a logical way to
investigate the possible mechanism of a catalyst that showed a high performance in
hydroprocessing of algal fuels. In this section, the HDN studies on algal bio-oils and
model chemicals were discussed together, which would give us new insights into the
catalyst development strategy.
4.1 Molybdenum based Catalysts
The most active catalyst of sulfided CoMo/Al2O3 was identified for the HDN and
HDO of algal bio-oils by researchers at the PNNL 47. The hydroprocessing was
conducted under relatively severe conditions in a continuous reactor. As the traditional
HDS catalysts for petroleum, CoMo/Al2O3 and NiMo/Al2O3, specifically sulfided form,
have been widely studied for HDS and HDN of model chemicals. These catalysts
exhibited higher HDS activity than HDN activity in competitive reactions between
thiophene and pyridine 102, and the presence of Co or Ni accelerated mainly the HDS
reaction 103. Because algal biofuels have low sulfur contents, the use of sulfided
17
catalysts will require the addition of external sulfur sources, e.g. hydrogen sulfide that
was able to enhance denitrogenation and inhibit hydrogenation 104. When applying the
sulfided catalyst in the batch reactor, sulfur is removed through the sulfiding process as
metal sulfides, poisoning catalysts 105. In addition, sulfided catalysts have a poor
hydrostability in the presence of water.
Instead of using sulfided CoMo or NiMo, unpromoted Mo 103, Mo sulfides 106,
nitrides 107, carbides, and phosphides 108 were proved to be more active in the HDN
reactions of model compounds than sulfided CoMo and NiMo.
The Mo nitrides (Mo2N) showed as much as 5-10 times more activities for pyridine
HDN than the sulfided Co-Mo/Al2O3 and MoS2/Al2O3 catalysts 109, and the selectivity
for C-N bond hydrogenolysis over C-C bond was higher for the nitride catalysts 110.
The bulk phase was predominantly γ-Mo2N with the surface consisting of either β-
Mo16N7 or mixtures of Mo and β-Mo16N7. The most active sites were located at the
perimeters of raft-like domains, while lower activity sites were associated with the γ-
Mo2N crystallite 111.
The Mo carbides (Mo2C) were proven to have the similar catalytic properties for
pyridine HDN to Mo nitrides 112. HDN of pyridine over Mo carbides and Mo nitrides
produced mostly cyclopentane and pentane, respectively. The selectivity difference
between Mo carbides and Mo nitrides might be due to differing bonding geometries for
pyridine on the Mo carbides and nitrides.
When MoP was tested for the catalytic activity in the HDN reaction of o-
propylaniline, the intrinsic HDN activity of the surface Mo atoms was about 6 times
higher than that of Mo edge atoms in MoS2/Al2O3 113, since the turnover numbers of
them were 13.610-4 molecules (Mo center) -1s-1 v.s. 2.210-4 molecules (Mo center) -
1s-1, respectively.
Further modification of Mo2C, Mo2N, and MoP could improve their activities.
Doping Mo2C with platinum (Mo2C-Pt) resulted in a higher HDN efficiency than Mo2C 114; nickel promoted Mo nitrides (NiMoNx with Ni 5 wt% & MoNx 15 wt%, supported
on γ-Al2O3) were more active than Mo2N 115; while addition of TiO2 to MoP/MCM-41
enhanced the C-N bond cleavage, but inhibited the dehydrogenation function 116.
Generally, the HDN activity of sulfided catalysts or Mo sulfides in the presence
of sulfur sources is always superior to other Mo compounds. However, due to the high
N content nature of algal fuels, Mo nitrides and carbides are more of interest to
hydroprocessing of algae-derived biofuels. Since the catalytic performance of Mo
nitrides and carbides replies on the surface and crystal structure 117, future research
attentions should be given to controlling crystal structure, surface modification, and
selecting suitable promoters and supports.
4. 2 Nobel Metal Catalysts
Noble metal catalysts (Ru, Pd, Rh, Ir, and Pt supported on carbon), specifically Ir
and Pt sulfides, exhibited higher pyridine HDN activity than the sulfided molybdenum-
based catalysts and can be used under milder conditions with high activity 118. When
hydrotreating algal HTL bio-crude, Pt/C and a mixture of Ru/C & Pt/γ-Al2O3 showed
good HDN activities, reducing the N content in hydrotreated oil to 1.5-1.8 wt% 23, 97.
18
But the activities of noble metal catalysts were reduced fast during hydroprocessing
algal bio-oils 95, the use of noble metals as the only active metal may not be suitable.
Pt and Ru were often used to modify other metal catalysts, such as tungsten carbides 119, Mo carbides 114, Fe 120, and CoMo 121. The HDN activity over Pt or Ru modified
catalysts was highly dependent on the amount of metallic sites introduced by them 119.
Although, noble metal catalysts are the most reactive metals for the C-N bond
cleavage, the cost and ease of deactivation are barriers for the process development.
The use of these metals for doping might improve the HDN activity of bimetallic
catalysts, as well as undergoing minimal hydrogenation reactions 122
4.3 Other Transitional Metal Catalysts
Nickel is an attractive metal in hydrotreating because of its high activity and low
cost 123. Ni has been used to modify many other metallic catalysts (like Mo and Mo2N),
and test results suggested that Ni-Mo species enhanced the hydrogenation of model
chemicals like pyridine 124. Catalytic hydroprocessing of algal bio-oil with Raney® Ni
led to the lowest N content of 1.6 wt% in upgraded oil 100. In addition, Ni phosphides
exhibited supreme efficiencies in HDS and HDN of model compounds, which were 99%
and 100%, respectively 125.
As a relatively new hydrogenation catalyst, iron (Fe) is considered as a low cost,
environmentally friendly, and sustainable material. However, the use of Fe as the only
active metal for HDN of algal oil resulted in a low conversion ratio and produced a
significant amount of alcohols 82. Instead, Fe doped Mo, tungsten (W) 126, and vanadium
(V) 127 sulfides were reported to give an unusual high HDN effect.
Tungsten carbides and phosphides were often used as HDN catalysts. One study
compared HDN of carbazole over W2C with Mo2C. The results indicated that W2C
possessed higher hydrogenation activity but lower total activity 128. Tungsten
phosphides were more extensively studied for their HDN behaviors 129. Bulk WP and
WP/SiO2 were found to be more active in HDN than W2C, W2N, WS2, and Ni-Mo-
S/Al2O3 catalysts 130.
Transition metal phosphides, such as WP, Ni2P, CoP, MoP, and Fe2P, emerged
recently as an attractive group of hydroprocessing catalysts, which have excellent
activity for HDS and HDN 131, 132. Study showed that their catalytic activities for
dibenzothiophene HDS and quinoline HDN followed the order: Ni2P> WP> MoP >
CoP> Fe2P. The crystal structure of metal phosphides is built with blocks of trigonal
prisms, which can well accommodate the large phosphorus atoms, leading to a more
isotropic crystal morphology and potentially better exposure of surface metal atoms to
fluid phase reactants 133. Furthermore, they show good heat and electricity
conductivities, and high thermal and chemical stability 134. However, the deactivation
of metal phosphides in the presence of water 135 and deficiency of P 125 was observed.
Because HTL and pyrolytic bio-oils might contain a higher amount of water and low
phosphorus, metal phosphides have less advantages in this application than metal
carbides and metal nitrates.
19
4.4 Effects of Supports
The catalyst support is the vehicle of active constituents, and affects the chemical
and physical properties of the catalyst, the degree of dispersion of active components,
and the stability. Catalyst supports also need to provide the certain reactants with high
surface area and suitable pore size. Currently, A12O3 was most widely used as the
support in traditional hydroprocessing catalysts, and studied for its effects on HDN
reactions. For instance, the alumina-supported molybdenum nitride (Mo2N/A12O3)
catalyst was extremely active in the hydrodenitrogenation of carbazole, compared to
the sulfided and reduced catalysts 136. The result indicated that the C-N hydrogenolysis
occurred on partially hydrogenated carbazole, suggesting the possibility of reducing
hydrogen consumption. A modification of the alumina support with borate ions could
increase the amount of acidity centers (or the acidity) of NiMo catalyst, leading to an
increased resistance to coking 137. However, it’s generally accepted that the Al2O3 was
not stable in the presence of large amount of water due to the formation of hydrated
boehmite 138, 139, and its acid sites could result in carbon deposition 140.
Compared with the A12O3 support, carbon supports showed better ability in water-
resistance and anti-coking. Hydroprocessing of pyridine over carbon-supported NiMo
sulfide formed less undesired products than Al-supported CoMo and NiMo catalysts 103. When the mesoporous carbon black support was employed to support Mo carbides,
the β-Mo2C hexagonal compact crystallographic phase was obtained as the unique
active phase, improving HDN of indole 141. The carbon-supported catalysts also showed
high resistance to poisoning 142, and high hydrodenitrogenation activity/selectivity 143.
Silica (e.g. MCM-41 116, SBA-15 86, 144) materials are of interests because of their
moderate acidity, high surface area, large pore size, and highly ordered structures. For
example, the SiO2 supported Mo or W phosphides showed superior HDN but lower
HDS activity compared to the sulfides 145. In addition, silica-alumina 146, 147, TiO2 148,
ZrO2 149, and CeO2-ZrO2
150 were also tested as supports in hydrotreating catalysts.
To summarize this subsection, a good HDN catalyst support should improve allover
thermal and chemical stability and the dispersity of active components, tailor surface
chemistry (for example, the HDN performance was related to the Brønsted acidity of
some catalysts 147, 151), and promote the formation of the highly active crystal structure.
4.5 Catalyst Suppliers
Hydroprocessing of algal fuels is a new and very challenging task. Most studies
were conducted by using commercial available catalysts. Table 5 gives a list of catalyst
suppliers, whose catalysts have been used for hydrotreating algal fuels. The catalysts
used for hydroprocessing lignocellulosic bio-oils and their suppliers can be found in 89.
20
Table 5. Catalyst Suppliers
Company name Catalysts used by
researchers
Company web
address
Reference
Sigma-Aldrich Pt/C, Pd/C, Ru/C, Mo2C,
MoS2, Al, Ni/SiO2-Al2O3,
HZSM-5, Al/Ni, and
activated carbon
sigmaaldrich.co
m
98, 100
Zeolyst International Zeolite Beta, ZSM-5, Zeolite
Y
zeolyst.com 98
Akzo Chemicals Inc. KF-1001 akzonobel.com 91
Alfa Aesar CoMo/γ-Al2O3 alfa.com 100
Johnson Matthey
(London, UK)
CoMo/Al2O3 and
NiMo/Al2O3
matthey.com 66
Qilu petrochemical
catalyst plant
(China)
NiMo/Al2O3 qpec.cn 25
5. Effect of Process Parameters on Hydroprocessing
The ultimate goal of hydroprocessing of algal biofuels is to synthesize drop-in fuels:
automobile fuels (gasoline and diesel) and aviation turbine fuels 152. These fuels are a
mixture of different hydrocarbons: The hydrocarbons of gasoline contain typically 4-
12 carbon atoms; diesel contains between 12 and 20 carbon atoms per molecule 153; and
the jet fuel has a carbon number distribution between about 8 and 16 154. Besides the
catalyst, the parameters of a hydroprocessing process include the reactor configuration,
reaction temperature, initial H2 pressure, residence time, and etc. These parameters are
important to the overall effectiveness of hydrogenation and the product distribution.
5.1 Crude Extracted Algal Oil
Normally, laboratory experiments were performed at the batch mode in either a
tubular reactor or a high pressure reactor, like autoclaves and Parr reactors 155. In one
case, the algal oil was treated sequentially in a two-reactor system with Pt/C and Pd/US-
Y zeolite, respectively, giving a 95% yield of alkanes 75. The advantages of using two-
stage reactors are that each reactor could be operated under different optimal conditions
for different catalysts, thus potentially giving a higher hydrocarbon yield.
In terms of reaction time, most experiments were conducted for 6-8 h, and then the
catalysts were deactivated due to coking. Only one report showed that when Pd/C was
used as the catalyst, their operation could last for 200-h if the algal oil was charged at
the rate of 0.177 mL/min 75. The long catalyst life in this study might be due to the
continuous operation mode, the use of two-stage reactors, and/or the catalyst support of
carbon.
Compared with other parameters, the product distribution is more likely
determined by the reaction temperature and the amount of initial H2 (i.e. the severity of
21
reaction conditions). Most hydroprocessing processes, which were conducted between
260-360°C with an initial H2 pressure of 2-5.5 MPa, successfully upgraded the algal
lipids to the diesel range (C13-C20). The major alkanes in treated oil have a carbon
number range of C15-C18. When a more severe condition of (400°C and 20 MPa H2)
was applied, the hydrocracking process was able to convert the algal lipids mainly to
gasoline (C4-C12), representing 67% of total hydrocarbons 80.
5.2 Bio-Oil
Some successful hydroprocessing experiments for upgrading bio-crude oils were
conducted in a two-stage continuous system 90. Both the continuous operation and the
two-stage treatment are important to the quality of the upgraded oil. For example, if the
batch reactors were used for the two-stage configuration, the N content of treated oils
was still above 2.4% 95.
Hydroprocessing of bio-oils was often performed for less than 24 h, and the life of
catalysts has not been studied systematically. The typical treatment temperature was
350-405°C, while the initial H2 pressure was around 6 MPa. In order to achieve a good
performance, a 10 MPa initial H2 pressure may be necessary 92, 93. Most studies were
able to obtain the upgraded oil with a carbon number distribution mainly in the diesel
range (C14-C18), representing 50-85% of total hydrocarbons 156.
6. Summary
The algal biofuel technology has been accelerated during last decade, especially
since 2010. However, in order to commercialize algae-based biofuels, it still requires
extensive efforts. The purpose of this review is to bring more research interests,
engineers, and catalysis scientists into this field via summarizing and criticizing the
state-of-the art in hydroprocessing of algae-based biofuels or biofuel precursors.
From a prospect of the process development, the continuous operation is highly
recommended, which might minimize the window that by-products react with the
upgraded oil. If it’s possible, the multi-stage continuous system is preferred, because
each reactor will be operated under different optimal conditions and/or different
catalysts. The typical reaction conditions for hydroprocessing of crude algal oil and bio-
crude oil are (260-360°C & 3-20 MPa H2) and (350-405°C & 6-13.6 MPa H2),
respectively. To obtain better HDO and HDN results, a higher initial H2 pressure (i.e.
the availability of H2) is expected.
From the point of view of catalyst development, the traditional HDS catalysts could
be efficient enough for deoxygenation of crude algal oils that have low sulfur and
nitrogen contents. Meanwhile, an ideal catalyst for hydroprocessing of algal bio-oils
should possess high activities towards both denitrogenation and deoxygenation.
According to Pacific Northwest National Laboratory's reports, the presulfided catalyst
of CoMo supported on fluorinated γ-Al2O3 was suggested to be the best candidate for
this process, if a continuous operation can be applied. However, the function of this
catalyst was only confirmed in a limited number of applications, and it did not show
22
the same efficiency in the batch reactor. Accordingly, the door is still open to catalysis
scientists who are interested in developing effective and cost-effective catalysts.
Up to date, it’s shown that bimetallic catalysts could be a promising choice to fulfill
the requirement for upgrading algal bio-oils that contain a high amount of nitrogenated
chemicals. The active metals of tungsten carbide (WC), molybdenum carbide (Mo2C),
and molybdenum nitride (MoN) are recommended, while the noble/transition metals of
Ru, Pt, Ni, Co, Fe, and Cu could be used to modify the active metal. Because the bio-
oil contains 5-10% moisture and hydrotreatment will produce water as a by-product,
the supporting material of the catalysts should be more water-resistance. Therefore, the
materials, such as carbon, modified Al2O3, and modified silica, are of interest.
For the long term, following issues shall to be considered: 1) The study on reaction
mechanism using model compounds is essential to reveal the catalysis pathway. 2) The
expected catalyst life is 2 years. Meanwhile catalysts need to tolerate poisons (such as
sulfur, phosphorus, and water), and minimize leaching problems and coke formation.
3) The economics for preparing catalysts are important, so the cost of active metals and
regeneration protocols are import factors.
Acknowledgments
We are grateful for the support from the School of Chemical Engineering and
Pharmacy at Wuhan Institute of Technology.
Reference
1. R. E. H. Sims, W. Mabee, J. N. Saddler and M. Taylor, Bioresource Technology, 2010, 101, 1570-
1580.
2. B. Antizar-Ladislao and J. L. Turrion-Gomez, Biofuels, Bioproducts and Biorefining, 2008, 2, 455-
469.
3. Y. Chisti, Biotechnology Advances, 2007, 25, 294-306.
4. L. Christenson and R. Sims, Biotechnology Advances, 2011, 29, 686-702.
5. M. Picardo, J. de Medeiros, J. Monteiro, R. Chaloub, M. Giordano and O. de Queiroz Fernandes
Araújo, Clean Techn Environ Policy, 2013, 15, 275-291.
6. NAABB, National Alliance for Advanced Biofuels and Bioproducts Synopsis,